CN115019531B - Vehicle control method and vehicle - Google Patents
Vehicle control method and vehicle Download PDFInfo
- Publication number
- CN115019531B CN115019531B CN202210612833.8A CN202210612833A CN115019531B CN 115019531 B CN115019531 B CN 115019531B CN 202210612833 A CN202210612833 A CN 202210612833A CN 115019531 B CN115019531 B CN 115019531B
- Authority
- CN
- China
- Prior art keywords
- lane
- lane change
- vehicle
- change position
- control method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000013178 mathematical model Methods 0.000 claims abstract description 21
- 230000014509 gene expression Effects 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 11
- 238000011478 gradient descent method Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 description 9
- 239000013598 vector Substances 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/0969—Systems involving transmission of navigation instructions to the vehicle having a display in the form of a map
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Navigation (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a vehicle control method and a vehicle, wherein the method comprises the following steps: acquiring a common navigation planning path on a common navigation map of a vehicle, and acquiring a stop lane change ending point and a target lane on a high-precision map according to the common navigation planning path; backtracking and searching each local lane changing termination point, a lane corresponding to each local lane changing termination point and variable lane distances on the corresponding lane according to the cut-off lane changing termination point on the high-precision map; judging whether the lane where the current vehicle is located is a target lane or not: if yes, directly adopting a common navigation planning path; if not, calculating the recommended optimal lane change position by using a lane change mathematical model according to the parameter space. The invention solves the technical problems that the navigation in the prior art is inaccurate, the reminding is not in time, the voice broadcasting or the road level guidance is insufficient to enable a driver to quickly understand the lane change reminding, and the safety risk of lane change is increased.
Description
Technical Field
The invention relates to the field of vehicles, in particular to a vehicle control method and a vehicle.
Background
In the prior art, navigation changes a channel through voice broadcasting, but the conditions of inaccuracy, untimely reminding and the like exist, and the user experience is not high; meanwhile, lane change reminding in the prior art is only based on related information of a map, consideration of driving related factors is not integrated, and requirements of drivers in different styles cannot be met when prompting is carried out; in the actual driving process, the conditions of untimely lane change reminding and unreasonable guidance often occur, the success rate of lane change is reduced, and the risk in the lane change process is increased; when the speed of the vehicle is too high or congestion occurs, voice broadcasting or road level guidance is insufficient to enable a driver to quickly understand lane change reminding, so that the safety risk of lane change is increased.
Disclosure of Invention
Based on the problems, the invention provides a vehicle control method and a vehicle, which solve the technical problems that navigation in the prior art is inaccurate, prompt is not timely, voice broadcasting or road level guidance is insufficient to enable a driver to quickly understand lane change prompt, and the safety risk of lane change is increased. The vehicle control method provided by the invention can timely and accurately calculate the optimal lane change point and timely remind the driver of the position of the optimal lane change point.
The invention provides a vehicle control method, which comprises the following steps:
acquiring a common navigation planning path on a common navigation map of a vehicle, and acquiring a stop lane change ending point and a target lane on a high-precision map according to the common navigation planning path;
backtracking and searching each local lane changing termination point, a lane corresponding to each local lane changing termination point and variable lane distances on the corresponding lane according to the cut-off lane changing termination point on the high-precision map;
judging whether the lane where the current vehicle is located is a target lane or not:
if yes, directly adopting a common navigation planning path;
if not, calculating a recommended optimal lane change position by using a lane change mathematical model according to a parameter space, wherein the parameter space comprises at least the following parameters: the method comprises the steps of changing the lane distance on a lane corresponding to each local lane changing termination point, judging the actual lane changing distance according to a high-precision map, and obtaining adjacent lane vehicle information of a lane where a current vehicle is located, wherein the adjacent lane vehicle information at least comprises the average speed of vehicles of the adjacent lanes and the average distance between the vehicles of the adjacent lanes;
the variable-channel mathematical model is a polynomial model, the polynomial model is the sum of a plurality of single expressions, the single expression containing parameters is a parameter single expression, the sum of indexes of one or more than two parameters in the parameter single expression is smaller than or equal to the highest degree of the polynomial, the parameters of the parameter single expression are randomly selected from a parameter space, and the polynomial model contains all parameter polynomials corresponding to the randomly selected parameters.
In addition, the method also comprises the step of updating the lane change mathematical model: and acquiring the actual lane change position of the vehicle, and acquiring an optimal set of weight coefficients by using a cost function through a gradient descent method according to the difference value between the actual lane change position and the recommended optimal lane change position so as to update the corresponding weight coefficients in the lane change mathematical model.
In addition, the optimization function is 1/2 Σ (y-y x)/(2), where y is the recommended optimal lane change position and y is the actual lane change position.
After the recommended optimal lane change position is calculated, the optimal lane change position is presented by the sound device and/or the display device of the vehicle.
Furthermore, calculating the recommended optimal lane change position using the lane change mathematical model based on the parameter space includes:
the parameter space is Θ= { θ 1 ,θ 2 ,θ 3 ,…,θ m M is the number of parameters in the parameter space;
the recommended optimal lane change position y is calculated according to the following lane change mathematical model:
wherein W is j (1. Ltoreq.j.ltoreq.m) is θ j The number of times of (2) is 0.ltoreq.W j ≤N，K n Representing the coefficient set, K n Take on the value { K 0 ,K 1 ,K 2 ,…,K n N is the highest degree of the polynomial.
The invention also provides a vehicle comprising the vehicle control method.
The invention solves the technical problems that the navigation in the prior art is inaccurate, the reminding is not in time, the voice broadcasting or the road level guidance is insufficient to enable a driver to quickly understand the lane change reminding, and the safety risk of lane change is increased. The vehicle control method provided by the invention can timely and accurately calculate the optimal lane change point and timely remind the driver of the position of the optimal lane change point.
Drawings
FIG. 1 is a flow chart of a method of controlling a vehicle according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of lane change of a vehicle in a vehicle control method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating lane change of a vehicle in a vehicle control method according to an embodiment of the present invention;
FIG. 4 is a schematic illustration of a prompt in a vehicle control method according to an embodiment of the present invention;
FIG. 5 is a schematic illustration of a prompt in a vehicle control method according to an embodiment of the present invention;
FIG. 6 is a schematic illustration of a prompt in a vehicle control method according to an embodiment of the present invention;
fig. 7 is a flowchart of a vehicle control method according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the specific embodiments and the accompanying drawings. It is intended that the invention be limited only by the specific embodiments illustrated and not by any means, but that it is intended that the invention be limited only by the terms of the appended claims.
Referring to fig. 1, the present invention proposes a vehicle control method including:
step S001, acquiring a common navigation planning path on a common navigation map of the vehicle, and acquiring a stop lane change ending point and a target lane on a high-precision map according to the common navigation planning path;
step S002, backtracking and searching each local lane changing termination point, a lane corresponding to each local lane changing termination point and variable lane distances on the corresponding lane according to the cut-off lane changing termination point on the high-precision map;
step S003, judging whether the lane in which the current vehicle is located is a target lane: if yes, directly adopting a common navigation planning path;
if not, calculating a recommended optimal lane change position by using a lane change mathematical model according to a parameter space, wherein the parameter space comprises at least the following parameters: the method comprises the steps of changing the lane distance on a lane corresponding to each local lane changing termination point, judging the actual lane changing distance according to a high-precision map, and obtaining adjacent lane vehicle information of a lane where a current vehicle is located, wherein the adjacent lane vehicle information at least comprises the average speed of vehicles of the adjacent lanes and the average distance between the vehicles of the adjacent lanes;
the variable-channel mathematical model is a polynomial model, the polynomial model is the sum of a plurality of single expressions, the single expression containing parameters is a parameter single expression, the sum of indexes of one or more than two parameters in the parameter single expression is smaller than or equal to the highest degree of the polynomial, the parameters of the parameter single expression are randomly selected from a parameter space, and the polynomial model contains all parameter polynomials corresponding to the randomly selected parameters. The parameters are arbitrarily selected from the parameter space, and all selected cases or combined cases are exhausted.
In step S001, optionally, according to the normal navigation planning path of the vehicle, the coordinates (x_w1, y_w1), (x_w2, y_w2) … … of the ending lane ending point, through which the normal navigation planning path passes, in the world coordinate system are searched in the high-precision map.
In step S002, a backtracking method is adopted, after the stopping lane changing ending point is found on the high-precision map, starting from the stopping lane changing ending point, each local lane changing ending point and the lane corresponding to each local lane changing ending point, and the variable lane distances on the corresponding lanes are searched according to the longitudinal displacement, so that the lane information taking each local lane changing ending point as the end point is obtained.
In step S003, optionally, coordinates (Xwc, ywc) of the vehicle in the world coordinate system are obtained by using the vehicle-mounted positioning device, and whether the lane in which the vehicle is currently located is located in the target lane or not is determined according to the coordinate information of the vehicle, the target lane is the lane with variable lane, and then the lane change termination point of each lane and the variable lane distance of each lane are provided on the high-precision map.
If yes, directly adopting a common navigation planning path;
if not, calculating a recommended optimal lane change position by using a lane change mathematical model according to a parameter space, wherein the parameter space at least comprises: the method comprises the steps of changing the lane distance on a lane corresponding to each local lane changing termination point, judging the actual lane changing distance according to a high-precision map, and obtaining adjacent lane vehicle information of a lane where a current vehicle is located, wherein the adjacent lane vehicle information at least comprises the average speed of vehicles of the adjacent lanes and the average distance between the vehicles of the adjacent lanes; optionally, vehicle-mounted sensors are utilized to acquire adjacent lane vehicle information of a lane in which the current vehicle is located.
The variable track distance on the lane corresponding to each local lane change ending point is the distance between the local lane change ending point and the vehicle, and the actual variable track distance obtained by judging according to the high-precision map is the distance between the vehicle and the local lane change ending point of which the lane line is a solid line. The recommended optimal lane-change position is at a position in the actual lane-change distance.
Alternatively, the polynomial model may be of the form where s 1 ,s 2 ,V av ,s crowd For parameters in the parameter space, the polynomial highest degree is 2:
from the above, it can be seen that the single equation can be derived from the parameter s 1 ,s 2 ,V av ,s crowd The sum of the indices of the parameters may be 1 or equal to the highest number of times 2.
The method adopts corresponding parameter polynomials of all possible optional selected parameters in the model, calculates the recommended optimal lane change position y by utilizing the thought of mathematical polynomial fitting approximation, and solves the technical problem that in the prior art, a clear functional relation cannot be given between factors (measured values obtained from a vehicle-mounted sensor or a high-precision map, such as adjacent lane vehicle information and the like) to be considered and the optimal lane change position y, so that the optimal lane change position cannot be predicted.
For example: firstly, determining the highest degree N of a polynomial, then combining parameter spaces formed by all factors affecting the calculation of the recommended optimal lane change position y on the premise of not exceeding the highest degree N to generate a polynomial containing constant terms, primary terms, secondary terms, N times, and assigning a weight coefficient to each term in the generated polynomial, so as to form a weight coefficient set (representing the relation between each parameter term combination and the recommended optimal lane change position y), wherein the weight coefficient set can be optimized through a learning model;
the weight coefficient set is optimized through online learning, firstly, a recommended optimal lane change position y is calculated according to the mentioned polynomial model and the current observation state quantity (according to a sensor or a high-precision map), then the current observation state quantity and a lane change point y selected in the actual driving process of a driver are recorded, and at the moment, a group of observation values (groudtuth) and theoretical values calculated by the current model are obtained, namely a group of training samples are obtained. The idea of learning optimization is to expect the theoretical value calculated by the polynomial model to be as close as possible to the actual value, i.e., (y-y) minimum, and 1/2 Σ (y-y) 2 is constructed as an optimization function, so that an optimal set of weight coefficients is obtained by a gradient descent method, and thus more accurate lane change reminding can be given for drivers of different styles.
Finally, through the online training of successive samples, the differential reminding is realized according to different driving styles after a large number of training samples are trained.
Referring to fig. 2 and 3, a vehicle travel-away high speed will be described as an example:
(1) Typical application modes are:
1. generating a common navigation planning path according to the custom destination, and acquiring road connection points (N1, N2) on the common navigation planning path;
2. according to the road connection point, the common navigation map is matched with the high-precision map, the left side in fig. 2 is a common navigation planning path, wherein R1 is a path before lane change, R2 and R3 are paths after the lane change and travel along the original road respectively, the right side is a path of the matched high-precision map, more lane information is given in the high-precision map, L1, L2, L3, L4 and L5 are lane paths before lane change, L6, L7 and L8 are lane paths before lane change, and L9 and L10 are lane paths after lane change.
3. When the distance between the vehicle position and the final lane change point of the exit ramp is smaller than K kilometers, a high-precision map of K kilometers in front is called (K is a system preset value);
4. and (3) finding out each local lane change ending point and lanes corresponding to each local lane change ending point on the front K kilometer high-precision map by adopting a backtracking method, wherein the lane change ending points of the lanes L1, L2 and L3 are respectively S2, S4 and S8.
5. And carrying out lane-level guidance and distance reminding by combining the calculated recommended optimal lane change position y, the lane where the vehicle is located, the current position of the vehicle, the longitudinal distance between the vehicle and the lane change point and the like.
Such as: the distance between S1 and S2 on the L1 lane path is the longitudinal distance between the vehicle and the lane change point, the distance between S3 and S4 on the L2 lane path is the longitudinal distance between the vehicle and the lane change point, the distance between S5 and S6 on the L3 lane path is the longitudinal distance between the vehicle and the lane change point, and the solid line between S6 and S7 cannot change lanes. D. C, B and A correspond to the variable track regions of L1, L2, L3, respectively.
(2) The process of finding the variable track area is as follows:
1. finding out the corresponding lane ids, namely L9 and L10, according to the road ids of the exit ramp;
2. finding a successor lane connected with L9 and L10, namely L4 and L5 according to the lane-level connection relation of the high-precision map;
3. judging the leftmost lane, namely L4, in the preceding lanes connected with the ramp according to the lane id;
4. finding all lanes on the left side of L4, namely L1, L2 and L3 according to the lane-level connection relation of the high-precision map;
5. in the lane information of L1, L2, L3, the start point and the end point of the lane line as the dotted line portions are acquired, that is: l1: s1-s2;
L2：s3-s4；
L3：s5-s6、s7-s8。
the invention solves the technical problems that the navigation in the prior art is inaccurate, the reminding is not in time, the voice broadcasting or the road level guidance is insufficient to enable a driver to quickly understand the lane change reminding, and the safety risk of lane change is increased. The vehicle control method provided by the invention can timely and accurately calculate the recommended optimal lane change point and timely remind the driver of the position of the optimal lane change point.
In one embodiment, the method further comprises the step of updating the lane change mathematical model: and acquiring the actual lane change position of the vehicle, and acquiring an optimal set of weight coefficients by using a cost function through a gradient descent method according to the difference value between the actual lane change position and the recommended optimal lane change position so as to update the corresponding weight coefficients in the lane change mathematical model.
The idea of mathematical polynomial fitting approximation is adopted, namely the lane change mathematical model is a polynomial model.
For example: firstly, determining the highest degree N of a polynomial, then combining parameter spaces formed by all factors affecting the calculation of the recommended optimal lane change position y on the premise of not exceeding the highest degree N to generate a polynomial containing constant terms, primary terms, secondary terms, N times, and assigning a weight coefficient to each term in the generated polynomial, so as to form a weight coefficient set (representing the relation between each parameter term combination and the recommended optimal lane change position y), wherein the weight coefficient set can be optimized through a learning model;
the weight coefficient set is optimized through online learning, firstly, a recommended optimal lane change position y is calculated according to the mentioned polynomial model and the current observation state quantity (according to a sensor or a high-precision map), then the current observation state quantity and a lane change point y selected in the actual driving process of a driver are recorded, and at the moment, a group of observation values (groudtuth) and theoretical values calculated by the current model are obtained, namely a group of training samples are obtained. The idea of learning optimization is to expect the theoretical value calculated by the polynomial model to be as close as possible to the actual value, i.e., (y-y) minimum, and 1/2 Σ (y-y) 2 is constructed as an optimization function, so that an optimal set of weight coefficients is obtained by a gradient descent method, and thus more accurate lane change reminding can be given for drivers of different styles.
In one embodiment, the optimization function is 1/2 Σ (y-y)/(2), where y is the recommended optimal lane change position and y is the actual lane change position. The theoretical value calculated by the optimization function is as close as possible to the actual value.
In one embodiment, after the recommended optimal lane change position is calculated, the optimal lane change position is prompted by the vehicle's audio and/or display device. The driver is more directly informed by the prompt.
Optionally, the reminding mode may be:
taking the case that the vehicle is in the L1 lane and is planned to drive off at a high speed as an example,
if the road map formed by the lane change ending point does not contain the current vehicle position information, displaying basic navigation information, as shown in fig. 4, including but not limited to current lane information, lane line type, navigation guide information and the like;
the generated reminding information comprises, but is not limited to, actual road images which are displayed on a meter or acquired by utilizing a vehicle-mounted camera, and lane line information (lane line type, position and the like) and navigation information are drawn in the images in a rendering mode and the like to realize reminding, and can also comprise relevant voice and sound reminding and necessary vibration information reminding;
and determining a path for the vehicle to drive away from a high speed according to the calculated combination of the optimal lane change points y.
As shown in fig. 5, the system set alert distances for the single lane change termination point and the final change termination point may be different for safe driving.
As shown in fig. 6, on the basis of the basic navigation information, current position information, a track changing direction, a track changing guiding track, a number of a track changing ending point set by the system, and the like can be added.
In one embodiment, calculating the recommended optimal lane change position using the lane change mathematical model based on the parameter space comprises:
the parameter space is Θ= { θ 1 ,θ 2 ,θ 3 ,…,θ m M is the number of parameters in the parameter space;
the recommended optimal lane change position y is calculated according to the following lane change mathematical model:
wherein W is j (1. Ltoreq.j.ltoreq.m) is θ j The number of times of (2) is 0.ltoreq.W j ≤N，K n Representing the coefficient set, K n Take on the value { K 0 ,K 1 ,K 2 ,…,K n N is the highest degree of the polynomial.
The following is the polynomial construction process:
the polynomial summation of y can be represented in matrix form for parameter set K by a set of data samples n Optimizing and learning;
set the coefficient K n Written in vector form: definition vector k= [ K ] 0 ,K 1 ,K 2 ,…,K n ]；
In polynomial summation, each term after expansion takes the part containing the parameters, written in the form of the following vectors:
constructing 1/2 sigma (y-y x)/(2) as an optimization function, and obtaining an optimal set of weight coefficients by a gradient descent method:
Θ＝{θ 1 ,θ 2 ,θ 3 ,…,θ m m is the number of parameter taking spaces, the highest frequency is N, N is more than or equal to 1 and less than or equal to m,
the parameter optimization (learning model) here steps as follows:
the general formula of the learning model can be expressed as follows, pi ^ I.e., is the best strategy, where the best set of coefficients K is obtained,
the loss function is defined as follows, y is the model calculated value, y * For the purpose of a true value,
optimizing through gradient descent until convergence to obtain an optimal coefficient set K,
alpha is the learning rate.
Illustrating how the best lane change point is calculated:
the parameter space includes: distance s of changing track end point 1 Actual variable track distance s 2 Average speed V of adjacent lanes av And adjacent lane traffic jam distance s crowd 。
Θ＝{s 1 ,s 2 ,V av ,s crowd }，
The highest number of times n=2 is defined,
the lane change point position y is expanded according to the following formula:
the form after deployment is as follows:
the polynomial sum representation of y may be represented as a matrix form as follows:
wherein:
L＝[k 0 ,k 1 ,k 2 ,k 3 ,k 4 ,k 5 ,k 6 ,k 7 ,k 8 ,k 9 ,k 10 ,k 11 ,k 12 ,k 13 ,k 14 ]therefore:
the parameter optimization (learning model) steps are as follows:
optimizing through gradient descent until convergence to obtain an optimal coefficient set K,
wherein is the learning rate.
Fig. 7 provides a vehicle control method. In this embodiment, when the own vehicle is determined to be in the lane change lane, the normal navigation planning path is continuously adopted (i.e. the vehicle is driven along the guiding of the basic navigation module), and if the own vehicle is not in the lane change lane (i.e. the target lane), the high-precision map planning path is adopted.
The invention also proposes a vehicle comprising a vehicle control method as defined in any one of the preceding claims. The invention solves the technical problems that the navigation in the prior art is inaccurate, the reminding is not in time, the voice broadcasting or the road level guidance is insufficient to enable a driver to quickly understand the lane change reminding, and the safety risk of lane change is increased. The vehicle control method provided by the invention can timely and accurately calculate the optimal lane change point and timely remind the driver of the position of the optimal lane change point.
What has been described above is merely illustrative of the principles and preferred embodiments of the present invention. It should be noted that several other variants are possible to those skilled in the art on the basis of the principle of the invention and should also be considered as the scope of protection of the present invention.
Claims (6)
1. A vehicle control method characterized by comprising:
acquiring a common navigation planning path on a common navigation map of a vehicle, and acquiring a stop lane change ending point and a target lane on a high-precision map according to the common navigation planning path;
according to the stopping lane changing ending points, backtracking and searching each local lane changing ending point, a lane corresponding to each local lane changing ending point and variable lane distances on the corresponding lane;
judging whether the lane where the current vehicle is located is a target lane or not:
if yes, directly adopting a common navigation planning path;
if not, calculating a recommended optimal lane change position by using a lane change mathematical model according to a parameter space, wherein the parameter space comprises at least the following parameters: the method comprises the steps of changing the lane distance on a lane corresponding to each local lane changing termination point, judging the actual lane changing distance according to a high-precision map, and obtaining adjacent lane vehicle information of a lane where a current vehicle is located, wherein the adjacent lane vehicle information at least comprises the average speed of vehicles of the adjacent lanes and the average distance between the vehicles of the adjacent lanes;
the variable-channel mathematical model is a polynomial model, the polynomial model is the sum of a plurality of single expressions, the single expression containing parameters is a parameter single expression, the sum of indexes of one or more than two parameters in the parameter single expression is smaller than or equal to the highest degree of the polynomial, the parameters of the parameter single expression are randomly selected from a parameter space, and the polynomial model contains all parameter polynomials corresponding to the randomly selected parameters.
2. The vehicle control method according to claim 1, characterized in that,
the method also comprises the steps of updating the lane change mathematical model: and acquiring the actual lane change position of the vehicle, and acquiring an optimal set of weight coefficients by using a cost function through a gradient descent method according to the difference value between the actual lane change position and the recommended optimal lane change position so as to update the corresponding weight coefficients in the lane change mathematical model.
3. The vehicle control method according to claim 2, characterized in that,
the optimization function is 1/2 Σ (y-y x) 2, where y is the recommended optimal lane change position and y is the actual lane change position.
4. The vehicle control method according to claim 1, characterized in that,
after the recommended optimal lane change position is calculated, the optimal lane change position is prompted by a sound device and/or a display device of the vehicle.
5. The vehicle control method according to any one of claims 1 to 4, characterized in that,
calculating the recommended optimal lane change position according to the parameter space by using a lane change mathematical model comprises:
the parameter space is Θ= { θ 1 ,θ 2 ,θ 3 ,…,θ m M is the number of parameters in the parameter space;
the recommended optimal lane change position y is calculated according to the following lane change mathematical model:
wherein W is j For theta j The times of (1) j is less than or equal to m, and 0 is less than or equal to W j ≤N，K n Representing the coefficient set, K n Take on the value { K 0 ,K 1 ,K 2 ,…,K n N is the highest degree of the polynomial.
6. A vehicle characterized by comprising employing the vehicle control method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210612833.8A CN115019531B (en) | 2022-05-31 | 2022-05-31 | Vehicle control method and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210612833.8A CN115019531B (en) | 2022-05-31 | 2022-05-31 | Vehicle control method and vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115019531A CN115019531A (en) | 2022-09-06 |
| CN115019531B true CN115019531B (en) | 2024-03-22 |
Family
ID=83071651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210612833.8A Active CN115019531B (en) | 2022-05-31 | 2022-05-31 | Vehicle control method and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115019531B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120071750A (en) * | 2010-12-23 | 2012-07-03 | 팅크웨어(주) | Electronic apparatus, method and recording medium for providing lane-changing information |
| CN105043403A (en) * | 2015-08-13 | 2015-11-11 | 武汉光庭信息技术有限公司 | High precision map path planning system and method |
| JP2017211301A (en) * | 2016-05-26 | 2017-11-30 | 本田技研工業株式会社 | Route guide device and route guide method |
| DE102018107508A1 (en) * | 2017-03-31 | 2018-10-04 | Ford Global Technologies, Llc | Real-time lane change indicator |
| GB2562376A (en) * | 2017-03-31 | 2018-11-14 | Ford Global Tech Llc | Virtual Steerable Path |
| CN108958242A (en) * | 2018-06-25 | 2018-12-07 | 武汉中海庭数据技术有限公司 | A kind of lane change decision assistant method and system based on high-precision map |
| CN110160552A (en) * | 2019-05-29 | 2019-08-23 | 百度在线网络技术（北京）有限公司 | Navigation information determines method, apparatus, equipment and storage medium |
| CN110597245A (en) * | 2019-08-12 | 2019-12-20 | 北京交通大学 | Automatic driving track-changing planning method based on quadratic planning and neural network |
| EP3750765A1 (en) * | 2019-06-14 | 2020-12-16 | Bayerische Motoren Werke Aktiengesellschaft | Methods, apparatuses and computer programs for generating a machine-learning model and for generating a control signal for operating a vehicle |
| CN112124310A (en) * | 2019-06-24 | 2020-12-25 | 初速度（苏州）科技有限公司 | Vehicle path transformation method and device |
| CN113071493A (en) * | 2021-04-16 | 2021-07-06 | 阿波罗智联(北京)科技有限公司 | Method, apparatus, storage medium, and program product for vehicle lane change control |
| CN113954837A (en) * | 2021-11-06 | 2022-01-21 | 交通运输部公路科学研究所 | Deep learning-based lane change decision method for large-scale commercial vehicle |
| CN113961002A (en) * | 2021-09-09 | 2022-01-21 | 浙江零跑科技股份有限公司 | Active lane change planning method based on structured road sampling |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007127419A (en) * | 2005-10-31 | 2007-05-24 | Aisin Aw Co Ltd | System and method for route guidance |
| US11458974B2 (en) * | 2019-07-08 | 2022-10-04 | Toyota Motor Engineering & Manufacturing North America, Inc. | Fleet-based average lane change and driver-specific behavior modelling for autonomous vehicle lane change operation |
-
2022
- 2022-05-31 CN CN202210612833.8A patent/CN115019531B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120071750A (en) * | 2010-12-23 | 2012-07-03 | 팅크웨어(주) | Electronic apparatus, method and recording medium for providing lane-changing information |
| CN105043403A (en) * | 2015-08-13 | 2015-11-11 | 武汉光庭信息技术有限公司 | High precision map path planning system and method |
| JP2017211301A (en) * | 2016-05-26 | 2017-11-30 | 本田技研工業株式会社 | Route guide device and route guide method |
| DE102018107508A1 (en) * | 2017-03-31 | 2018-10-04 | Ford Global Technologies, Llc | Real-time lane change indicator |
| GB2562376A (en) * | 2017-03-31 | 2018-11-14 | Ford Global Tech Llc | Virtual Steerable Path |
| CN108958242A (en) * | 2018-06-25 | 2018-12-07 | 武汉中海庭数据技术有限公司 | A kind of lane change decision assistant method and system based on high-precision map |
| CN110160552A (en) * | 2019-05-29 | 2019-08-23 | 百度在线网络技术（北京）有限公司 | Navigation information determines method, apparatus, equipment and storage medium |
| EP3750765A1 (en) * | 2019-06-14 | 2020-12-16 | Bayerische Motoren Werke Aktiengesellschaft | Methods, apparatuses and computer programs for generating a machine-learning model and for generating a control signal for operating a vehicle |
| CN112124310A (en) * | 2019-06-24 | 2020-12-25 | 初速度（苏州）科技有限公司 | Vehicle path transformation method and device |
| CN110597245A (en) * | 2019-08-12 | 2019-12-20 | 北京交通大学 | Automatic driving track-changing planning method based on quadratic planning and neural network |
| CN113071493A (en) * | 2021-04-16 | 2021-07-06 | 阿波罗智联(北京)科技有限公司 | Method, apparatus, storage medium, and program product for vehicle lane change control |
| CN113961002A (en) * | 2021-09-09 | 2022-01-21 | 浙江零跑科技股份有限公司 | Active lane change planning method based on structured road sampling |
| CN113954837A (en) * | 2021-11-06 | 2022-01-21 | 交通运输部公路科学研究所 | Deep learning-based lane change decision method for large-scale commercial vehicle |
Non-Patent Citations (1)
| Title |
|---|
| 基于驾驶人不满度的高速公路自动驾驶换道决策;陈慧;王洁新;;中国公路学报(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115019531A (en) | 2022-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8583366B2 (en) | Road shape learning apparatus | |
| US10479362B2 (en) | Autonomous driving assistance system, autonomous driving assistance method, and computer program | |
| US10399571B2 (en) | Autonomous driving assistance system, autonomous driving assistance method, and computer program | |
| US10300916B2 (en) | Autonomous driving assistance system, autonomous driving assistance method, and computer program | |
| CN110641480B (en) | Automatic driving function pushing method and system based on traffic flow and vehicle | |
| US8005602B2 (en) | Vehicle speed control device, method of determining target speed by using the device, and program executing the method | |
| CN110182209A (en) | Controller of vehicle and control method for vehicle | |
| CN109844669B (en) | Vehicle control device | |
| US6873911B2 (en) | Method and system for vehicle operator assistance improvement | |
| US20190084579A1 (en) | Autonomous driving assistance device and computer program | |
| US20090240426A1 (en) | Navigation device and navigation method | |
| JP4776476B2 (en) | Navigation device and method for drawing enlarged intersection | |
| JP3378490B2 (en) | Road information recognition device | |
| JPH10122884A (en) | Navigator | |
| JP2002329299A (en) | Curve approach controller | |
| JPH11208386A (en) | Vehicular motion control device | |
| JP2011203053A (en) | Drive assist system | |
| WO2021049108A1 (en) | Vehicle control device, vehicle control method, vehicle motion control system, and lane estimation device | |
| EP2546608B1 (en) | Map display system, method, and program | |
| US20190227552A1 (en) | Vehicle control device | |
| JP5565561B2 (en) | Driving assistance system | |
| CN115019531B (en) | Vehicle control method and vehicle | |
| CN112512885A (en) | Driving assisting method and device and automobile | |
| JP5881426B2 (en) | Tunnel information use navigation device | |
| JP2018063658A (en) | Travel plan generating apparatus and computer program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |